So, is the bubble chart the best way to show the data? Does the interactivity help? How can it be improved?

I pulled the data from this Google spreadsheet provided by OpenSecrets.org in a write-up posted yesterday.

The post Donor Bubbles appeared first on HelpMeViz.

We’ve been struggling with how to make this table reader friendly for the non-data literate.  Please help!  This table is included in a larger report that will be distributed to community agencies with varying levels of data literacy. Please provide help using Excel—we don’t have access to other software programs (unless they are free!).

Data are available here.

The post Table Design Help appeared first on HelpMeViz.

I am aiming to compare two categories (both in percentage), one of which can and does have negative values for particular locations, and then grouped into regions. There is also a 75% threshold that I would like to show – the 75% is relevant only for the blue category.

What I hope to show is the relationship between the blue (% variance to budget) and orange (% variance FTEs) categories. For example, when the % variance to budget values exceed the 75% threshold–which is good–is that because the %FTEs are positive percentages (i.e., they used more staff than budgeted)? Or does the % variance to budget exceed 75% (again, the good result) when fewer staff are budgeted (negative % FTE)? While the current graph has the two bars side by side is there a better way to show the relationship? Any other general observations would also be appreciated as I don’t know how to avoid having some location labels overlapping the negative bars.

The Excel file is available here.

The post Percentages and Targets appeared first on HelpMeViz.

I’ve built an open-source Chrome extension that replaces Chrome’s default history page with a visually oriented interface. It captures and displays in Pinterest-style format all the pages you’ve visited and it let’s you search the content of all saved pages.

For privacy and performance, the extension stores everything locally using the browser’s built-in NoSQL store (IndexedDB) and it can store up to 10,000 pages including screenshots and full text, which comes to about 5GB of disk space, but this limit is arbitrary. It will be possible for the user to set a much lower or much higher value in the next version.

The extension is capable of recording the time you spend each day on each URL (how long in total the URL is loaded in a visible tab/window excluding idle time) and it can therefore tell you things like how much time you’ve spent on Hacker News or Gmail over any period. However, the visualizations can be a lot more powerful and meaningful if the user opts in to share their anonymized browsing history. With that, we can tell you if you’re spending too much time on Hacker News or on the D3 mailing list, as compared to the total population of users.

It can do another thing by gathering anonymized browsing data, per opt-in permission, which is to use that to draw correlations between seemingly unrelated sites, which users can visualize, may be in a network graph where edges grow thicker or thinner depending on link strength and nodes grow bigger or smaller depending on popularity of the given site among all users, but it can be a chord diagram, too. I’m not a visualization expert.

The extension can also monitor things like rapid back and forth switching between open tabs, where the sites are not correlated at all, as it may indicate ADHD behavior (but that’s just an idea that I personally find fascinating, although it may be something that users don’t wish to analyze — where to draw the line…)

Given all of the potential for Personal Web Analytics, the type of input I’m asking for is all about the imagination, not so much concrete ideas. I just want to hear what people think and what kind and variety of personal web analytics they would find useful.

There is also some Hacker News discussion of the existing, non-data-visualization features of the extension, where, for improved privacy, I promised to add a “Do Not Capture” list of hostnames (e.g. mail.google.com) that users can declare. This feature has been added in the latest version.

The post All Seeing Eye Chrome Extension appeared first on HelpMeViz.

The post SNAP Participation & Outreach appeared first on HelpMeViz.

The visualization is a clustered bar chart that shows the correlation between the percent change in payroll and performance between 2002 and 2012. There are two bars on the y-axis: the thick light blue one represents the percent change in payroll between 2002 and 2012, and the thin dark bar represents the change in percent wins between 2002 and 2012. The grey and red bars are for negative values

In an iteration to the graph, I rearranged the bars following the winning percentage in the overall League. (Data are here.)

The post MLB Payroll and Wins appeared first on HelpMeViz.

The post Urban, Population, Agriculture, and Land appeared first on HelpMeViz.

The post Plotting Internet Use appeared first on HelpMeViz.

I’m trying to figure out how to create “nice” panel charts in excel, where I’m currently attempting to do so in Stata. (See for example Figure 2.) I can easily do the individual excel chart, but my investigation online into making this a 2X2 set seems to involve using pivot tables, and even then I’m not quite convinced it will work out as I want it to. I wanted to reach out in the event that this is something you or others have done before – Excel trend figures side by side for comparison purposes…also known to some as panel charts. This allows the reader to view trend lines across different groupings over the same scale to make helpful comparisons.

The data are available here.

Here is the Excel file Ira suggested.

The post Income Volatility Line Graphs appeared first on HelpMeViz.

Thanks,
Jon

The post Trying to Fix “The Fix” appeared first on HelpMeViz.